40results about How to "Realize automatic design" patented technology

The invention discloses a flight program design system for performance-based navigation (PBN), a verification platform and a verification method, belonging to the field of aerial navigation. The program design system is based on an AIXM (Aeronautical Information Exchange Model) core database; a program protection region can be automatically drawn, an obstacle is evaluated, and a chart is generated; and thus, functions such as seamless connection and the like of the PBN program protection region and the conventional flight program protection region can be realized. According to the system, the chart can be automatically generated according to data finished by the flight program design, customized edition and modification can also be realized according to requirement, and program design and publishing are finished at one time. By using the flight program design system, the verification platform and the verification method, the PBN program can be directly added in a navigation database of the verification platform, and the navigation database can be edited and modified in a verification process. In addition, a visual scene database is increased; the defects and the disadvantages of the program design can be directly discovered according to geomorphology in a visual scene when the PBN flight program is verified; and the program is further modified perfectly.

A method for evaluate that comprehensive performance of a supercritical carbon dioxide cycle power generation system of a small lead-based reactor. Firstly, the basic operating parameters of the supercritical carbon dioxide (S-CO 2) cycle power generation system for lead-based reactor are determined. Subsequently, the modular design method is used to establish the sub-modules of turbomachinery work, heat transfer, heat exchanger pressure drop calculation, heat exchanger efficiency calculation, thermal analysis and economic analysis of power generation system. Then, the iterative calculation method is used to solve the heat exchanger pressure drop calculation and the thermodynamic analysis sub-module of the power generation system, and the thermal parameters of the power generation system and the unit power cost are obtained. Finally, based on multi-objective collaborative optimization, the thermal performance of power generation system is evaluated. Economy. This method can effectivelycomplete the new type of small-sized lead-based reactor S_Sco2 Design of Cycle Power Generation System, Thermodynamic, Heat Transfer and Economic Analysis Models for CO2 Cycle Power Generation Systems, Efficient System Heat Transfer-Economic performance analysis, solve the problem of automatic system design, complete the comprehensive performance comparison and guide the optimal S-CO2 Cycle PatternScreening.

The invention discloses a system for automatically designing a radiotherapy plan based on machine learning. The system comprises an input unit, a radiation field parameter predicting unit, an optimization target and constrained condition predicting unit, and an inverse plan optimization unit, wherein the input unit is used for acquiring an original image of the patient and segmented interested area information. The radiation parameter predicting unit is used for constructing a learning model based on a neural network and automatically predicting the radiation field parameter. The optimizationtarget and constrained condition predicting unit is used for constructing a dosage distribution predicting model based on a neural network, and automatically converts the predicted expected dosage distribution to a target function and the constrained condition required for inverse optimization. The inverse plan optimization unit is used for performing optimization according to the target function,the constriction setting and the radiation field parameter by means of an optimization method for obtaining the sub-field which corresponds with each radiation field direction and the weight thereof,thereby finishing planned designing. The system has advantages of realizing automatic plan designing, greatly reducing workload of a plan designer and improving working efficiency.

The invention provides a method for designing an optical free-form surface optical system based on point-by-point computation. The method comprises the following steps: constructing an initial plane system according to design parameters, wherein the initial plane system comprises: a primary mirror, a secondary mirror and a third mirror; selecting characteristic light; performing fitting based on aselected characteristic curve, calculating free-form surface data points on the third mirror, performing fitting, in which two factors, namely, normal and coordinates are considered, by using a four-order XY polynomial so as to correct the initial structure of the third mirror; performing fitting based on the selected characteristic curve, calculating the free-form surface data points on the primary mirror, and performing fitting, in which two factors, namely, normal and coordinates are considered, by using the four-order XY polynomial so as to correct the initial structure of the primary mirror; and performing iterative optimization by using the system in which the third mirror and the primary mirror are corrected as the initial structure of the iteration process, and obtaining the free-form surface optical system.

The invention discloses a language description-based house type design method, device and system. The method comprises the steps of obtaining voice data of a user and converting the voice data into text data; extracting attribute information about house type design in the text data, taking the attribute information as nodes, connecting the nodes according to text semantics, and forming an attribute semantic graph; carrying out layout prediction on the attribute semantic graph by adopting a layout prediction model constructed based on a neural network, outputting a predicted house type layout graph, and carrying out optimization processing on the house type layout graph to obtain an optimized house type layout graph; generating texture information according to the text semantics by adoptinga texture generation model constructed based on a generative adversarial network; and generating a three-dimensional scene model according to the optimized house type layout diagram and the texture information, and rendering to obtain a multi-view scene rendering diagram. According to the house type design method, device and system, automatic house type design is carried out according to the voice data of the user, and the house type design efficiency is improved on the basis that the designed house type meets the user requirement.

The invention discloses an automatic design method for a control law in a mode conversion process of a self-adaptive cycle engine, which is used for automatically calculating the control law in the mode conversion process after working condition parameters and the control law of a starting point and an ending point of the mode conversion process are given. The mode conversion design is simplifiedinto research of two key working points from a section of working process, the particle swarm optimization algorithm is combined with the sensitivity analysis method to solve the control rules of thekey working points, then the control rules of the whole mode conversion process are obtained, manual adjustment is not needed in the control rule solving process, and the working efficiency is improved. Design efficiency can be greatly improved.

The invention relates to the technical field of computer vision, and relates to a method and device for searching a coordinate recognition model based on an NAS technology, in particular to a method and device for training the coordinate recognition model, and the method comprises the steps: obtaining the value range of each hyper-parameter of a baseline model, and obtaining the value range of each hyper-parameter according to the value range of each hyper-parameter; combining the hyper-parameters under different values to generate a plurality of numerical value combinations; for each numerical value combination, setting each hyper-parameter of the baseline model as each value in any numerical value combination, and obtaining a candidate coordinate identification model under the numerical value combination; for each candidate coordinate recognition model, inputting the image sample set into any candidate coordinate recognition model for training, and calculating an error value of the candidate coordinate recognition model; and taking the candidate coordinate recognition model meeting the preset error value condition as a final optimized coordinate recognition model, so that automatic design of the coordinate recognition model can be realized in combination with NAS.

The invention provides a flow chart generation method and device, equipment and a storage medium. The method comprises the steps of obtaining a flow description text of a user, carrying out the word segmentation, obtaining a plurality of first vocabularies, carrying out the part-of-speech tagging, recognizing each part-of-speech tagging through a preset intention recognition model, obtaining an intention feature vector of each first vocabulary, obtaining an intention vector corresponding to each sentence through a preset method, identifying a flow dimension of each sentence, generating a corresponding flow frame, and further generating a flow chart. The method has the advantages that automatic design of the flow chart is achieved, and compared with manual design of the flow chart, a large amount of time cost and labor cost are saved.

The embodiment of the invention provides an interface color setting method, device and equipment and a storage medium, and the method comprises the steps: receiving a color setting instruction of a user, wherein the color setting instruction comprises a color parameter of a target color set by the user; generating a color plate according to the color parameters of the target colors and a preset color plate generation rule, wherein the color plate comprises the target colors and adjacent colors of the multiple target colors, and the adjacent colors are lighter and/or deeper than the target colors; and setting the color of each object in the interface according to the color palette and the mapping relationship between each object in the preset interface and each color position in the color palette. In the embodiment of the invention, in the interface design and development process, a user can obtain a plurality of other adjacent colors only by determining the target color to generate thecolor palette, automatic design of the interface color can be realized according to the color palette and the mapping relationship between each object in the preset interface and each color positionin the color palette, the color standard of interface design can be unified, the interface design and development efficiency is improved, and human resources are saved.

The invention discloses a manufacturing method and a device of a special tooling for tilting tile, relating to the technical field of compressors, which mainly aims at improving the accuracy of the design of the special tooling for the tilting tile, and improving the manufacturing efficiency and the manufacturing accuracy of the special tooling for the tilting tile. The method comprises: by usingthe special tooling of the tilting tile block of the existing model, when the parameter information of the new type tilting tile is received, the special tooling 3D model corresponding to the new typetilting tile is generated according to the parameter information of the new type tilting tile and the parameterized tooling 3D model. A manufacture scheme of a special tooling correspond to a new type of tilting tile is determined by utilizing a three-dimensional model of that special tooling and a manufacturing module provide by a preset modeling software, and the manufacturing scheme is uploaded to a turning-milling composite machining center to manufacture a special tooling corresponding to the new type of tilting tile.

The invention relates to a highway route constructing method. The highway route constructing method comprises the following steps: (1), establishing a virtual design route for a highway to be constructed, wherein the highway to be constructed consists of multiple design routes; (2), according to a starting stake mark, a finishing stake mark and a broken chain of each design route, establishing stake mark interval tables of a left half and a right half of the virtual design route; (3), setting stake mark identifiers of the virtual design route; (4), setting stake mark identifiers of the left half and the right half of the virtual design route; (5), converting the stake mark identifiers of the virtual design route into stake mark identifiers of a real design route; and (6), finishing design on the true design route. By the highway route constructing method, survey and design steps are greatly simplified and the design efficiency and the design quality are improved.

The embodiment of the invention provides a manipulator design method and device, a medium and an electronic device, and the method comprises the steps: obtaining a machine learning model and model parameters thereof, and the model parameters of the machine learning model being obtained through the training of the machine learning model through the historical feature information and historical feedback information of a historical manipulator; constructing an objective function according to the machine learning model and model parameters thereof; determining constraint conditions of the currentfeature information of the current manipulator according to the demand information of the target object for the current manipulator; and obtaining design information of the current manipulator according to the target function and the constraint condition. According to the technical scheme, the design of the current manipulator can be assisted based on the historical manipulator information.